Zobrazeno 1 - 10
of 315
pro vyhledávání: '"Goldowsky A"'
Autor:
Goldowsky, Howard, Sarathy, Vasanth
We propose an approach to analogical inference that marries the neuro-symbolic computational power of complex-sampled hyperdimensional computing (HDC) with Conceptual Spaces Theory (CST), a promising theory of semantic meaning. CST sketches, at an ab
Externí odkaz:
http://arxiv.org/abs/2411.08684
Autor:
Balesni, Mikita, Hobbhahn, Marius, Lindner, David, Meinke, Alexander, Korbak, Tomek, Clymer, Joshua, Shlegeris, Buck, Scheurer, Jérémy, Stix, Charlotte, Shah, Rusheb, Goldowsky-Dill, Nicholas, Braun, Dan, Chughtai, Bilal, Evans, Owain, Kokotajlo, Daniel, Bushnaq, Lucius
We sketch how developers of frontier AI systems could construct a structured rationale -- a 'safety case' -- that an AI system is unlikely to cause catastrophic outcomes through scheming. Scheming is a potential threat model where AI systems could pu
Externí odkaz:
http://arxiv.org/abs/2411.03336
Identifying the features learned by neural networks is a core challenge in mechanistic interpretability. Sparse autoencoders (SAEs), which learn a sparse, overcomplete dictionary that reconstructs a network's internal activations, have been used to i
Externí odkaz:
http://arxiv.org/abs/2405.12241
Autor:
Bushnaq, Lucius, Heimersheim, Stefan, Goldowsky-Dill, Nicholas, Braun, Dan, Mendel, Jake, Hänni, Kaarel, Griffin, Avery, Stöhler, Jörn, Wache, Magdalena, Hobbhahn, Marius
Mechanistic interpretability aims to understand the behavior of neural networks by reverse-engineering their internal computations. However, current methods struggle to find clear interpretations of neural network activations because a decomposition
Externí odkaz:
http://arxiv.org/abs/2405.10928
Autor:
Bushnaq, Lucius, Mendel, Jake, Heimersheim, Stefan, Braun, Dan, Goldowsky-Dill, Nicholas, Hänni, Kaarel, Wu, Cindy, Hobbhahn, Marius
Mechanistic Interpretability aims to reverse engineer the algorithms implemented by neural networks by studying their weights and activations. An obstacle to reverse engineering neural networks is that many of the parameters inside a network are not
Externí odkaz:
http://arxiv.org/abs/2405.10927
Localizing behaviors of neural networks to a subset of the network's components or a subset of interactions between components is a natural first step towards analyzing network mechanisms and possible failure modes. Existing work is often qualitative
Externí odkaz:
http://arxiv.org/abs/2304.05969
Autor:
Claudia Sampaio da Silva, Julia Alicia Boos, Jonas Goldowsky, Manon Blache, Noa Schmid, Tim Heinemann, Christoph Netsch, Francesca Luongo, Stéphanie Boder-Pasche, Gilles Weder, Alba Pueyo Moliner, Roos-Anne Samsom, Ary Marsee, Kerstin Schneeberger, Ali Mirsaidi, Bart Spee, Thomas Valentin, Andreas Hierlemann, Vincent Revol
Publikováno v:
Frontiers in Bioengineering and Biotechnology, Vol 12 (2024)
End-stage liver diseases have an increasing impact worldwide, exacerbated by the shortage of transplantable organs. Recognized as one of the promising solutions, tissue engineering aims at recreating functional tissues and organs in vitro. The integr
Externí odkaz:
https://doaj.org/article/22da439e01c74baf80d340ea0048f0d2
Autor:
Sarah Heub, Fatemeh Navaee, Daniel Migliozzi, Diane Ledroit, Stéphanie Boder-Pasche, Jonas Goldowsky, Emilie Vuille-Dit-Bille, Joëlle Hofer, Carine Gaiser, Vincent Revol, Laura Suter-Dick, Gilles Weder
Publikováno v:
Scientific Reports, Vol 12, Iss 1, Pp 1-9 (2022)
Abstract Standardised and high-throughput methods have been developed for the production and experimental handling of some 3D in vitro models. However, adapted analytical tools are still missing for scientists and researchers to fully exploit the pot
Externí odkaz:
https://doaj.org/article/67b79a7869044e69b3482be18cc2959e
Autor:
GOLDOWSKY, BARBARA
Publikováno v:
Chicago Review, 2021 Jan 01. 64/65(4/1), 230-237.
Externí odkaz:
https://www.jstor.org/stable/27191767
Publikováno v:
In SLAS Technology October 2018 23(5):470-475